{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TXHETB3TX6E5Q75QA4IX53Q3VO","short_pith_number":"pith:TXHETB3T","schema_version":"1.0","canonical_sha256":"9dce498773bf89d87fb007117eee1bab92937d15b4f7bee1fa793101b6cae2ff","source":{"kind":"arxiv","id":"2302.05508","version":2},"attestation_state":"computed","paper":{"title":"FairPy: A Toolkit for Evaluation of Prediction Biases and their Mitigation in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hrishikesh Viswanath, Tianyi Zhang","submitted_at":"2023-02-10T20:54:10Z","abstract_excerpt":"Recent studies have demonstrated that large pretrained language models (LLMs) such as BERT and GPT-2 exhibit biases in token prediction, often inherited from the data distributions present in their training corpora. In response, a number of mathematical frameworks have been proposed to quantify, identify, and mitigate these the likelihood of biased token predictions. In this paper, we present a comprehensive survey of such techniques tailored towards widely used LLMs such as BERT, GPT-2, etc. We additionally introduce Fairpy, a modular and extensible toolkit that provides plug-and-play interfa"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2302.05508","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-10T20:54:10Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"20ef2f305a20a71c9f5305ff602e1a8ab2de5d7fc1c7a6b5f6bbae26cf547cc7","abstract_canon_sha256":"4816965e24bc92d08cd10f63f61cc7068bac2d136ffc8cb95c2eb694e89050c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:49:02.405795Z","signature_b64":"IRhhXmwloPz4TYOLiw/pg2fI5Uo43O/Cgo4qlA3i1Xa9DRqs5Nix4E3MU4vq6IevyjvRZATdCKhiBimy8sTwAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9dce498773bf89d87fb007117eee1bab92937d15b4f7bee1fa793101b6cae2ff","last_reissued_at":"2026-07-05T10:49:02.405290Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:49:02.405290Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FairPy: A Toolkit for Evaluation of Prediction Biases and their Mitigation in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hrishikesh Viswanath, Tianyi Zhang","submitted_at":"2023-02-10T20:54:10Z","abstract_excerpt":"Recent studies have demonstrated that large pretrained language models (LLMs) such as BERT and GPT-2 exhibit biases in token prediction, often inherited from the data distributions present in their training corpora. In response, a number of mathematical frameworks have been proposed to quantify, identify, and mitigate these the likelihood of biased token predictions. In this paper, we present a comprehensive survey of such techniques tailored towards widely used LLMs such as BERT, GPT-2, etc. We additionally introduce Fairpy, a modular and extensible toolkit that provides plug-and-play interfa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.05508","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2302.05508/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2302.05508","created_at":"2026-07-05T10:49:02.405361+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.05508v2","created_at":"2026-07-05T10:49:02.405361+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.05508","created_at":"2026-07-05T10:49:02.405361+00:00"},{"alias_kind":"pith_short_12","alias_value":"TXHETB3TX6E5","created_at":"2026-07-05T10:49:02.405361+00:00"},{"alias_kind":"pith_short_16","alias_value":"TXHETB3TX6E5Q75Q","created_at":"2026-07-05T10:49:02.405361+00:00"},{"alias_kind":"pith_short_8","alias_value":"TXHETB3T","created_at":"2026-07-05T10:49:02.405361+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TXHETB3TX6E5Q75QA4IX53Q3VO","json":"https://pith.science/pith/TXHETB3TX6E5Q75QA4IX53Q3VO.json","graph_json":"https://pith.science/api/pith-number/TXHETB3TX6E5Q75QA4IX53Q3VO/graph.json","events_json":"https://pith.science/api/pith-number/TXHETB3TX6E5Q75QA4IX53Q3VO/events.json","paper":"https://pith.science/paper/TXHETB3T"},"agent_actions":{"view_html":"https://pith.science/pith/TXHETB3TX6E5Q75QA4IX53Q3VO","download_json":"https://pith.science/pith/TXHETB3TX6E5Q75QA4IX53Q3VO.json","view_paper":"https://pith.science/paper/TXHETB3T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.05508&json=true","fetch_graph":"https://pith.science/api/pith-number/TXHETB3TX6E5Q75QA4IX53Q3VO/graph.json","fetch_events":"https://pith.science/api/pith-number/TXHETB3TX6E5Q75QA4IX53Q3VO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TXHETB3TX6E5Q75QA4IX53Q3VO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TXHETB3TX6E5Q75QA4IX53Q3VO/action/storage_attestation","attest_author":"https://pith.science/pith/TXHETB3TX6E5Q75QA4IX53Q3VO/action/author_attestation","sign_citation":"https://pith.science/pith/TXHETB3TX6E5Q75QA4IX53Q3VO/action/citation_signature","submit_replication":"https://pith.science/pith/TXHETB3TX6E5Q75QA4IX53Q3VO/action/replication_record"}},"created_at":"2026-07-05T10:49:02.405361+00:00","updated_at":"2026-07-05T10:49:02.405361+00:00"}